plot.error {randomSurvivalForest} | R Documentation |
Plot the out-of-bag (OOB) error rates for the ensemble CHF. Also plot variable importance (VIMP) for predictors. This is the default plot method for the package.
plot.error(x, sorted = TRUE, ...) plot.rsf(x, sorted = TRUE, ...)
x |
An object of class (rsf, grow) or (rsf,
predict) . |
sorted |
Should variables be sorted by importance values? |
... |
Further arguments passed to or from other methods. |
Plot of OOB error rate, with the b-th value being the error rate for the ensemble computed using the first b trees. Error rate is 1-C, where C is Harrell's concordance index. Rates given are between 0 and 1, with 0.5 representing the benchmark value of a procedure based on random guessing. A value of 0 is perfect.
In the orginal call, if importance
=TRUE (the default setting),
then variable importance (VIMP) values for predictors are plotted. A
matrix with up to 3 columns is also printed. First column are VIMP
values, second column are standardized VIMP values (divided by the
maximum importance value), third column is the vector
predictorWt
(only printed if values are distinct).
For competing risks, error rates and VIMPs are given for the ensemble CHF and for the ensemble conditional CHF for each event type.
Hemant Ishwaran hemant.ishwaran@gmail.com and Udaya B. Kogalur ubk2101@columbia.edu
H. Ishwaran, U.B. Kogalur, E.H. Blackstone and M.S. Lauer (2008). Random survival forests, Ann. App. Statist., 2:841-860.
L. Breiman (2001). Random forests, Machine Learning, 45:5-32.
F.E. Harrell et al. (1982). Evaluating the yield of medical tests, J. Amer. Med. Assoc., 247, 2543-2546.
rsf
,
predict.rsf
.
## Not run: data(veteran, package = "randomSurvivalForest") v.out <- rsf(Survrsf(time, status)~., veteran, ntree = 1000, nsplit = 3) plot.error(v.out) ## End(Not run)